OBJECTIVE: To examine the potential significance of the sequence of the onset of dieting and binge eating in binge eating disorder (BED).
DESIGN: BED patients were interviewed and completed a battery of psychometrically well-established measures of current eating behaviors, eating disorder psychopathology, and associated psychological functioning.
SUBJECTS: Participants were 98 consecutive outpatients with BED evaluated for a clinical trial.
MEASURES: Interview data, self-report measures and measured body weight were examined.
RESULTS: Participants who reported that dieting preceded binge eating (DIETfirst, 65%) were compared to those who reported that binge eating preceded their first diet (BINGEfirst, 35%). The study groups did not differ in demography, current or highest body mass index, current eating behaviors or psychopathology, or psychological functioning. The two groups did not differ in age of first diet; however, the BINGEfirst group was significantly younger when first overweight, at onset of binge eating, and at onset of BED diagnosis. The BINGEfirst group reported a higher frequency of being teased about their weight.
CONCLUSIONS: A substantial subgroup of BED patients report that binge eating preceded their first diet. This finding, which replicates previous reports for BED and appears higher than that generally reported for bulimia nervosa, may have implication for etiologic models of binge eating.
Binge eating disorder (BED) is a new eating disorder category included in the DSM-IV in Appendix B reflecting ‘criteria sets provided for further study’. The multisite field trials of the proposed diagnostic criteria1,2 and subsequent studies have reported general, albeit mixed, support for the validity of the BED diagnosis (see Ref 3). Many questions remain regarding the development and psychopathology of BED.3,4,5
Restrictive dieting is central to most etiologic and risk models of eating disorders.6 Patients with eating disorders generally have extensive histories of dieting, and a period of restrictive dieting frequently precedes the onset of bulimia nervosa in the vast majority of cases.7 However, some investigators have reported that roughly 15% of patients with bulimia nervosa were not dieting at the onset of binge eating8,9 and that this subgroup may differ from the ‘typical cases’ with bulimia nervosa.9,10
The emerging empirical literature for BED, in contrast to that for bulimia nervosa, suggests that binge eating may precede dieting in many cases. Wilson and colleagues11 found that only 8.7% of obese binge eaters entering a weight control program reported that they had been on a strict diet at the time they began binge eating and that 64% reported that the binge eating preceded their obesity.
In a study of 30 females enrolled in a BED treatment program, 44% reported binge eating by age 16 whereas only 28% reported significant dieting and weight loss by age 16;12 binge eating preceded obesity by several years in the majority of cases. A recent study with 87 BED patients entering a clinical trial found that 55% of cases reported binge eating prior to dieting.13 These findings are generally consistent with those reported in the DSM-IV field trials.2
Given the centrality of dieting or restraint in etiologic models of bulimia nervosa7 and the emerging possibility of a potentially different sequence for binge eaters who do not purge,11,12,13 the current study aimed to examine the potential significance of the sequence of the onset of dieting or binge eating in BED.
Subjects were 98 adults (80 females and 18 males) consecutively evaluated for outpatient clinical trials who met DSM-IV criteria for BED. Subjects were aged 18–60 y (mean=41.6, s.d.=9.9), 90% (n=88) were Caucasian, 57% (n=56) were married, and 86% (n=84) either attended or graduated from college. Written informed consent was obtained from participants.
DSM-IV axis I psychiatric diagnoses were derived by consensus and based on the independent administration of the SCID-I14 and a clinical interview by trained and monitored Ph.D.-level research clinicians. DSM-IV axis II personality disorders were assigned based on the administration of the Diagnostic Interview for DSM-IV Personality Disorders (DIPD-IV),15 a semi-structured diagnostic interview that assesses all 12 (10 recognized and two research diagnoses) personality disorders.
The SCID-I and DIPD-IV interviews were performed by three experienced Ph.D.-level research clinicians with training in the SCID-I and eating disorders. The three SCID-I interviewers for this study were evaluated as part of a larger inter-rater reliability study involving 12 interviewers.16 Inter-rater reliability (calculated using 84 pairs of raters)—as reflected by kappa coefficient—for eating disorder diagnoses was 0.77 for all raters and was 1.0 for the BED diagnosis for the three interviewers in the present study. Kappas for the personality disorder diagnoses ranged from 0.58 to 1.0.16
The clinical interviews were performed by three additional Ph.D. research clinicians, including the two authors (CMG, RMM). The clinical interviews were performed using a structured format to obtain the historical variables of interest. Inter-rater reliability for these highly structured interviews was not assessed. The variables of principal interest here (ie, onset, sequence, etc) represent concrete variables rather than dimensional or clinical factors that require complex judgements for ratings. Nonetheless, the first author reviewed the administration of the interviews with the research clinicians on an on-going basis in an attempt to minimize potential drift.
Diagnoses were further confirmed by relevant portions of two self-report measures, the Eating Disorder Examination Questionnaire (EDE-Q17) and the Questionnaire For Eating and Weight Practices—Revised (QEWP-R4), which were administered to assess the history and current eating-related functioning of the participants.
The EDE-Q17 is the self-report version of the investigator-based Eating Disorder Examination interview.18 The EDE-Q assesses the frequency of the core behavioral features (ie objective binge episodes and subjective binge episodes) and intensity of eating-related psychopathology producing four subscales (ie restraint, eating concerns, weight concerns and shape concerns scales). The EDE-Q has been empirically validated with BED19,20 and BN.17,21
The QEWP-R,4 employed in the DSM-IV field trials,2 assesses each criterion of BED including the stipulated 6-month duration. The QEWP-R also assesses a number of relevant historical variables to this study, including age at which first overweight by 10 pounds and the number of 20-pound weight loss–regain cycles.
The clinical interviewer checked the relevant QEWP-R4 items with participants and also inquired about the following variables following a structured format: age of first diet, and age of onset for regular binge eating (once per week). The interviewer obtained actual weights using a medical balance beam scale and calculated body mass index (BMI; weight/ by height2). The independently administered SCID-I14 assessed for age of onset of BED diagnosis at threshold DSM-IV criteria (ie 2 days with binges per week for at least 6 months).
In addition, several psychometrically sound self-report measures of important associated domains were administered. These were selected based on conceptual and empirical grounds (see Ref 3) and to allow for comparison with previous work.9,12,13
(1) The Physical Appearance-Related Teasing Scale (PARTS22) is an 18-item measure of the frequency of being teased while growing up. The PARTS has two subscales: (a) weight/size teasing and (b) general appearance teasing. Higher scores reflect higher frequencies of reported teasing experiences. The PARTS is psychometrically sound,22 and has been used in nonclinical23,24 and clinical studies25 of obesity, eating problems and body image. Childhood teasing about weight/shape has been identified as a possible risk factor for BED.5,25
(2) The Body Shape Questionnaire (BSQ26) is a 34-item measure of body dissatisfaction. The BSQ assesses the frequency of preoccupation with and distress about body size/shape. Subjects rate items (eg ‘Have you felt so bad about your shape that you have cried?’) on a scale from 1 (never) to 6 (always); thus higher scores reflect greater body dissatisfaction. The BSQ is a widely used instrument in studies of eating and weight disorders.9,23,24,25
(3) The Three Factor Eating Questionnaire (TFEQ26) is a widely used psychometrically sound measure, with three subscales to tap important eating disordered domains: dietary restraint, disinhibition and hunger. The TFEQ is widely used in obesity and eating disorder studies.9,28
(4) The Beck Depression Inventory (BDI29) 21-item version was employed to assess level of depression. The BDI is a psychometrically sound, widely used inventory of the cognitive, affective, motivational and somatic symptoms of depression. Higher scores reflect higher levels of depression; a score of 16 or higher is generally recommended as a cutoff point for major depression.29,30
(5) The Rosenberg Self-Esteem Scale (RSE31) is a widely used 10-item measure of global self-esteem with established reliability and validity.31,32 Subjects rate the items (eg ‘On the whole, I am satisfied with myself’) on a scale from 1 (strongly agree) to 4 (strongly disagree). Scoring procedures result in higher, scores reflecting higher self-esteem.
(6) The Drug Abuse Screening Test (DAST33) is a 27-item self-report screening instrument for drug abuse and related adverse consequences experienced secondary to the drug use. The DAST has demonstrated reliability and validity and good predictive power for substance abuse problems in varied patient populations.33,34
(7) The Self-Consciousness Scale (SCS35) contains three scales, two of which were administered here: (a) the Public Self-Consciousness Scale contains seven items that measure awareness of others’ reactions to the self (eg ‘I'm concerned about what other people think of me’), and (b) the Social Anxiety Scale contains six items that measure level of discomfort experienced in the presence of others (eg ‘It takes me time to overcome my shyness in public situations’). Subjects rate the items on a scale from 0 (extremely uncharacteristic) to 4 (extremely characteristic of self). The SCS has demonstrated good reliability and validity.35 The SCS has been found to be associated with body dissatisfaction in bulimia nervosa36 and obesity37 and has been found to be higher in obese binge eaters than obese nonbinge eaters.37
Sixty-four of the 98 participants (65%) who reported that dieting preceded binge eating (DIETfirst group) were compared to the 34 participants (35%) who reported that binge eating preceded their first diet (BINGEfirst group). The DIETfirst and BINGEfirst groups did not differ in current age (41.1 y vs 42.7 y; F(1, 97)=0.57, ns) or gender (81% female vs 82% female; chi-square (1)=0.02, ns).
The two study groups were characterized by similar axis I psychiatric and axis II personality disorder frequencies. The two study groups did not differ in the distribution of major depression (χ21=0.47, ns), dysthymia (χ21=0.12, ns), panic disorder (χ2(1)=0.00, ns), social phobia (χ2(1)=1.49, ns), obsessive compulsive disorder (χ2(1)=0.41, ns), post-traumatic stress disorder (χ2(1)=0.19, ns), or substance use disorders (χ2(1)=0.88, ns). No differences in the distribution of any lifetime psychiatric disorders were observed. Complete axis II personality disorder data derived with the DIPD-IV were available for 74 of the participants. The two study groups did not differ significantly in whether any personality disorder was diagnosed (χ2(1)=0.29, ns) or in the distribution of specific cluster A (χ2(1)=0.01, ns), cluster B (χ2(1)=0.63), or cluster C (χ2(1)=1.56, ns) personality disorder diagnoses.
Table 1 summarizes the developmental variables, current eating disturbance, and current psychological functioning findings for the two study groups. Also shown in Table 1 are the statistical tests for significant differences (ANOVAs for continuous data). The DIETfirst and BINGEfirst groups did not differ in age of first diet; however, the BINGEfirst group was significantly younger when first overweight, at onset of binge eating, and at onset of BED diagnosis. The BINGEfirst group reported a higher frequency of being teased about their weight and size while growing up (PARTS).
The two groups were similar in current body mass index (BMI) and highest lifetime BMI, current binge eating frequency (objective and subjective binge eating frequency on the EDE-Q), associated eating-related psychopathology (EDE-Q and TFEQ), body image dissatisfaction (BSQ), and current psychological functioning (BDI, RSE, DAST and SCS).
We re-analyzed all of the data separately for females only (note that gender was not differentially distributed by onset of díet or binge eating as indicated above). When restricted to females (n=80), all of the overall nonsignificant findings remained nonsignificant. In the case of significant findings observed for the overall group, all of the comparison remained significant for females only. For females only, the BINGEfirst group was significantly younger when first overweight (M=11.2 (s.d.=6.1) vs M=16.7 (s.d.=8.4) y; F=8.79, P=0.004), at onset of binge eating (M=10.7 (s.d.=4.4) vs M=23.9 (s.d.= 10.8) y; F=36.72, P<0.000), and at onset of BED diagnosis (M=15.3 (s.d.=8.7) vs M=26.6 (s.d.= 11.5) y; F=11.9, P<0.001). For females, the BINGEfirst group reported a higher frequency on the PARTS of being teased about their weight and size while growing up (M=33.61 (s.d.=14.9) vs M=23.56 (s.d.=11.9); F=7.34, P<0.008).
This study rigorously assessed a consecutive series of patients with BED at a university-based outpatient eating disorder program. Sixty-five percent of patients reported that dieting preceded their binge eating and 35% reported that binge eating preceded their first diet. Age of onset of binge eating and of BED differed significantly depending on whether dieting or binge eating began first. Patients who began binge eating first reported earlier onset of overweight, higher frequency of being teased about their weight and shape, and an earlier onset of BED diagnosis than those patients who reported dieting prior to their first binge.
No significant differences between the two study groups were observed in demography or gender, current BMI or highest BMI, weight cycling, eating behavior disturbances (binge eating frequency, overeating frequency, hunger, restraint, disinhibition), overvalued ideas regarding weight and shape, body image dissatisfaction, associated psychological domains (depression, self-esteem, drug problems, or self-consciousness), or in the distribution of DSM-IV psychiatric or personality disorders.
Roughly one-third of BED patients reported regular binge eating prior to regular dieting. This finding, while slightly lower than reported previously for obese binge eaters11,12 and BED,13 is nonetheless considerably higher than the consistent finding that dieting precedes binge eating in the vast majority of bulimic cases.9,10
Overall for the entire BED study group, the average age of first overweight was 14.6 y, age of first diet was 16.2 y, and age of binge eating was 19.4 y. This sequence, at first glance, might be thought to be consistent with prevailing views that overweight leads to dieting which precipitates binge eating in vulnerable persons. However, 34% of the patients reported a sharp deviation from this pattern. For BED patients who report binge eating first, the mean age of onset of binge eating was 11.6, the mean age of first overweight was 12.4 y, and the mean age of first diet was 17.1 y. In this pattern, binge eating leads to overweight which leads to dieting. This finding and findings from other studies10 suggest the importance of early interventions for binge eating.
We found that BED patients who report binge eating first reported a higher frequency of being teased about weight and size than BED patients who report dieting first. This may be due to the finding that they were more likely to be overweight at an earlier age. Grilo and colleagues25 reported that the frequency of being teased about weight and size while growing up was positively correlated with body dissatisfaction during adulthood in obese female patients. Moreover, early onset obesity was associated with greater body dissatisfaction,25 a finding initially noted by Stunkard.38,39 Fairburn and colleagues,5 using a community-based case-control design, found that repeated exposure to negative comments about weight and shape emerged as a main risk factor that identified BED from healthy controls and from psychiatric controls. Compared to subjects with other psychiatric disorders, subjects with BED reported more childhood obesity and more teasing experiences.5 It is possible, however, that the weight-related teasing was due to childhood overweight but unrelated to the binge eating per se. This finding merits further research to better determine the sequence and potential etiologic significance. Longitudinal prospective designs40,41 with use of integrated and relevant comparison groups (eg Ref 5; see Ref 42) are needed to better establish significance and understand complex interplay of multiple variables.43
The lack of differences in current body mass index (BMI), and the striking similarities in current eating behaviors, attitudes regarding weight/shape and psychological functioning replicate those previously reported for BED.13 A similar report for bulimia nervosa9 indicated that 16 bulimics who began binge eating before dieting had higher body dissatisfaction and higher disinhibition than 81 bulimics who began binge eating after dieting; they too, however, failed to find differences in current eating and binge eating behaviors or depression.
These descriptive findings may have prognostic significance. Agras and colleagues44 reported that early onset binge eating predicted poor outcomes from psychological treatments otherwise characterized by impressive efficacy. Future descriptive studies of the psychopathology of BED and of treatment outcome should consider age of onset and sequence of dieting vs binge eating.
We observed no differences between the two groups in DSM-IV psychiatric or personality disorders. These findings also generally replicate those previously reported13 for most areas of DSM-III-R psychiatric functioning. Spurrell and colleagues13 reported that BED patients who had an onset of binge eating prior to dieting had significantly higher lifetime rates of substance use disorders. No other current or lifetime psychiatric disorders or the presence/absence of any personality disorder differed significantly between BED patients who binge or dieted first. Spurrell and colleagues,13 however, did note nonsignificant trends for the binge first group to have higher rates of current substance abuse and personality disorders.
We note several potential limitations. Our study group was recruited for clinical trials; thus, the findings may not be generalizable to general outpatient or community populations, or to BED individuals who self-select to weight control rather than eating disorder programs. It is possible that such persons may have different histories and variations in the development and presenting manifestations of BED. Moreover, we note our reliance on retrospective recall of age and sequence of diet and binge eating on both self-report and interview assessments. Retrospective recall may be inaccurate or influenced by systematic biases. We also relied on participants’ definitions of diets, which can vary across people. Longitudinal prospective studies are needed to determine developmental pathways.41
In conclusion, a substantial subgroup of BED patients (roughly one-third) report that binge eating preceded their first diet. This finding, which replicates previous reports for BED, and appears higher than that generally reported for bulimia nervosa, may have implications for etiologic models of binge eating. Further research, particularly multivariate43 and longitudinal40,41 work, is needed to understand the nature and significance of the possible differences in developmental pathways to BED.
Spitzer RL, Devlin M, Walsh BT, Hasin D, Wing RR, Marcus MD, Stunkard AJ, Wadden TA, Yanovski S, Agras WS, Mitchell J, Nonas C . Binge eating disorder: a multisite field trial of the diagnostic criteria Int J Eating Disord 1992 11: 191–203.
Spitzer RL, Yanovski S, Wadden T et al.Binge eating disorder: Its further validation in a multisite study Int J Eating Disord 1993 13: 137–153.
Grilo CM . The assessment and treatment of binge eating disorder J Pract Psychiat Behav Health 1998 4: 191–201.
Yanovski SZ . Binge eating disorder: current knowledge and future directions Obes Res 1993 1: 306–324.
Fairburn CG, Doll HA, Welch SL, Hay PJ, Davies BA, O'Connor ME . Risk factors for binge eating disorder Arch Gen Psychiat 1998 55: 425–432.
Brownell KD, Fairburn CG . Eating disorders and obesity: a comprehensive handbook Guilford Press: New York 1995.
Polivy J, Herman CP . Dieting and bingeing: a causal analysis Am Psychol 1985 40: 193–201.
Mitchell JE, Hatsukami D, Pyle RL, Eckert ED . The bulimia syndrome: Course of illness and associated problems Comp Psychiat 1986 27: 165–170.
Haiman C, Devlin MJ . Binge eating before the onset of dieting: a distinct subgroup of bulimia nervosa? Int J Eating Disord 1999 25: 151–157.
Mussell MP, Mitchell JE, Fenna CJ, Crosby RD, Miller JP, Hoberman HM . A comparison of onset of binge eating versus dieting in the development of bulimia nervosa Int J Eating Disord 1997 12: 353–360.
Wilson GT, Nonas CA, Rosenblum GD . Assessment of binge eating in obese patients Int J Eating Disord 1993 150: 1472–1479.
Mussell MP, Mitchell JE, Weller CL et al.Onset of binge eating, dieting, obesity, and mood disorders among subjects seeking treatment for binge eating disorder Int J Eating Disord 1995 17: 395–401.
Spurrell EM, Wilfley DE, Tanofsky, Brownell KD . Age of onset for binge eating: are there different pathways to binge eating? Int J Eating Disord 1997 21: 55–65.
First MB, Spitzer RL, Gibbon M, Williams JBW . Structured clinical interview for DSM-IV Axis I disorders—patient edition (SCID-I/P, Version 2.0). NYSPI 1996.
Zanarini M . The diagnostic interview for DSM-IV personality disorders Harvard Medical School: Harvard, MA 1996.
Zanarini MC, Skodol AE, Bender D, Dolan R, Sanislow C, Schaeffer E, Morey LC, Grilo CM, Shea TM, McGlashan TH, Gunderson JG . The Collaborative Longitudinal Personality Disorders Study: reliability of Axis I and Axis II diagnoses. J Personality Disord (in press).
Fairburn CG, Beglin SJ . Assessment of eating disorders: Interview or self-report questionnaire? Int J Eating Disord 1994 16: 363–370.
Fairburn CG, Cooper Z . The Eating Disorder Examination (12th edition). In CG Fairburn, GT Wilson (eds) Binge eating: nature, assessment, and treatment Guilford Press: New York 1993 pp. 317–360.
Wilfley DE, Schwartz MB, Spurrell EB, Fairburn CG . Assessing the specific psychopathology of binge eating disorder patients: interview or self-report? Behav Res Ther 1997 12: 1151–1159.
Grilo CM, Masheb RM, Daniels ES, Berman R, Wilson GT . Validation study of the Eating Disorder Examination—Questionnaire in binge eaters World Congress of behavioral and cognitive therapies Acapulco, Mexico, July 1998.
Loeb KL, Pike KM, Walsh BT, Wilson GT . Assessment of diagnostic feature of bulimia nervosa: interview versus self-report format Int J Eating Disord 1994 16: 75–81.
Thompson JK, Fabian LJ, Moulton DO, Dunn ME, Altabe MN . Development and validation of the physical appearance related teasing scale J Person Assess 1991 56: 513–521.
Beren SE, Hayden HA, Wilfley DE, Grilo CM . The influence of sexual orientation on body image in adult men and women Int J Eating Disord 1996 20: 135–141.
Akan GE, Grilo CM . Sociocultural influences on eating attitudes and behaviors, body image, and psychological functioning: a comparison of African-American, Asian-American, and Caucasian women Int J Eating Disorder 1995 18:: 181–187.
Grilo CM, Wilfley D, Brownell KD, Rodin J . Teasing, body image, and self-esteem in a clinical sample of obese women Addictive Behav 1994 19: 443–450.
Cooper PJ, Taylor MJ, Cooper Z, Fairburn CG . The development and validation of the Body Shape Questionnaire Int J Eating Disord 1987 6: 485–494.
Stunkard AJ, Messick S . The three factor eating questionnaire to measure dietary restraint, disinhibition, and hunger. J Psychosomatic Res 1985 29: 71–83.
Allison DB, Kalinsky LB, Gorman BS . A comparison of the psychometric properties of three measures of dietary restraint Psychol Assessment 1992 4: 391–398.
Beck AT, Steer RA . Manual for revised Beck Depression Inventory Psychological Corporation: New York 1987.
Beck AT, Rush AJ, Shaw BF, Emery G . Cognitive therapy of depression Guilford Press: New York 1979.
Rosenberg M . Conceiving the self Basic Books: New York 1979.
Griffiths RA, Beumont PJV, Giannakopoulos E, Russell J, Schotte D, Thornton C, Touyz SW, Varano P . Measuring self-esteem in diet disordered patients: the validity of the Rosenberg and Coopersmith contrasted Int J Eating Disord 1999 25: 227–231.
Skinner HA . The Drug Abuse Screening Test Addict Behav 1992 7: 363–371.
Martino S, Grilo CM, Fehon D . The development of the Drug Abuse Screening Test for Adolescents Addict Behav 2000 25: 57–70.
Fenigstein A, Scheier MF, Buss AH . Public and private self-consciousness: assessment and theory J Consult Clin Psychol 1975 43: 522–527.
Striegel-Moore RH, Silberstein LR, Rodin J . The social self in bulimia nervosa: public self-consciousness, social anxiety, and perceived fraudulence J Abnorm Psychol 1993 102: 297–303.
Grilo CM, Wilfley DE, Jones A, Brownell KD, Rodin J . The social self, body dissatisfaction, and binge eating in obese females Obes Res 1994 2: 24–27.
Stunkard AJ, Mendelson M . Obesity and the body image: I. Characteristics of disturbances in the body image of some obese persons Am J Psychiat 1967 123: 1296–1300.
Stunkard AJ, Burt V . Obesity and the body image: II. Age at onset of disturbances in the body image Am J Psychiat 1967 123: 1443–1447.
Stice E, Mazotti L, Krebs M, Martin S . Predictors of adolescent dieting behaviors: a longitudinal study Psychol Addict Behav 1998 12: 195–205.
Stice E, Agras WS . Predicting onset and cessation of bulimic behaviors during adolescence: a longitudinal grouping analysis Behav Ther 1998 29: 257–276.
Allison DB . A note on the selection of control groups and control variables in comorbidity research Comp Psychiat 1993 34: 336–339.
Friedman MA, Brownell KD . Psychological correlates of obesity: moving to the next research generation Psychol Bull 1995 117: 3–20.
Agras S, Telch C, Arnow B, Eldredge K, Henderson J, Marnell M . Does interpersonal therapy help patients with binge eating disorder who fail to respond to cognitive-behavioral therapy? J Consult Clin Psychol 1995 63: 356–360.
This work was supported by NIH grant DK49587 (Dr. Grilo). We thank Dr. Elayne Daniels for her contribution to the assessment of participants
About this article
A Pilot Study Evaluating a Binge-Eating Screener in Children: Development of the Children’s Brief Binge-Eating Questionnaire in a Pediatric Obesity Clinic
Clinical Pediatrics (2019)
European Eating Disorders Review (2017)
Binge Eating and Weight Loss Outcomes in Individuals with Type 2 Diabetes: 4-Year Results from the Look AHEAD Study
Clinical Psychology Review (2016)